Numerous models for supervised and reinforcement learning benefit from
combinations of discrete and continuous model components. End-to-end learnable
discrete-continuous models are compositional, tend to generalize better, and
are more interpretable. A popular approach to building discrete-continuous
computation graphs is that of integrating discrete probability distributions
into neural networks using stochastic softmax tricks. Prior work has mainly
focused on computation graphs with a single discrete component on each of the
graph's execution paths. We analyze the behavior of more complex stochastic
computations graphs with multiple sequential discrete components. We show that
it is challenging to optimize the parameters of these models, mainly due to
small gradients and local minima. We then propose two new strategies to
overcome these challenges. First, we show that increasing the scale parameter
of the Gumbel noise perturbations during training improves the learning
behavior. Second, we propose dropout residual connections specifically tailored
to stochastic, discrete-continuous computation graphs. With an extensive set of
experiments, we show that we can train complex discrete-continuous models which
one cannot train with standard stochastic softmax tricks. We also show that
complex discrete-stochastic models generalize better than their continuous
counterparts on several benchmark datasets.

通过增加 Gumbel 噪声扰动的规模参数和引入特定于随机离散 - 连续计算图的 dropout 残差连接，可以训练复杂的离散 - 连续模型，且其泛化能力优于连续模型。

离散连续计算图的高效学习

Efficient Learning of Discrete-Continuous Computation Graphs

The Gumbel-Max trick is the basis of many relaxed gradient estimators. These
estimators are easy to implement and low variance, but the goal of scaling them
comprehensively to large combinatorial distributions is still outstanding.
Working within the perturbation model framework, we introduce stochastic
softmax tricks, which generalize the Gumbel-Softmax trick to combinatorial
spaces. Our framework is a unified perspective on existing relaxed estimators
for perturbation models, and it contains many novel relaxations. We design
structured relaxations for subset selection, spanning trees, arborescences, and
others. When compared to less structured baselines, we find that stochastic
softmax tricks can be used to train latent variable models that perform better
and discover more latent structure.

在扰动模型框架下，我们使用随机 softmax 技巧（stochastic softmax tricks）对 Gumbel-Softmax 技巧进行了推广，将其推广到组合空间。与基线相比，我们发现使用随机 softmax 技巧可以训练表现更好、发现更多潜在结构的潜在变量模型。